
Challenge overview
The Grounded CoMPaT Recognition (GCR) is a compositional 3D vision task that aims to collectively recognize and ground compositions of materials on parts of 3D objects. This task is based on the 3DCoMPaT dataset, a large-scale dataset composed of stylized 3D objects and associated 2D renderings. We propose two variations of this task: GCR-Coarse and GCR-Fine, which are based on coarse-grained and fine-grained 3D segmentations of the 3DCoMPaT models. We highly encourage participants of the challenge to enter and submit to both tracks of the challenge.π Dataset
The 3DCoMPaT dataset for both challenge tracks is available through our download page.π¨ Submission
Submission will be made through the eval.ai platform.π Rules
Here are the rules for the challenge:- Submission Limit: Each participant is allowed to submit their solution a maximum of three times per day.
- Data Usage: Participants are not permitted to use any data other than the 3DCoMPaT data for training their models.
- Technical Report: Each participant must submit a technical report detailing their methods, which will be made public, in order to be eligible for any prizes or rewards.
π Awards
Total prize pool: $3000. Teams are encouraged to particpate to both challenge tracks.Fine track:
- 1st: $1300
- 2nd: $700
Coarse track:
- 1st: $700
- 2nd: $300